Adsorption Kinetics and Breakthrough of Carbon Dioxide for the Chemical Modified Activated Carbon Filter Used in the Building

Abstract

We studied different face velocity and carbon dioxide (CO₂) initial concentration to examine the adsorption behavior of calcium oxide (CaO) impregnated activated carbon (AC) filter and also to discuss pseudo-first-order, pseudo-second-order and intra-particle diffusion three kinetic models. The experimental results show that saturation time and saturation capacity were decreased and increased with higher inlet concentration at the same face velocity, respectively. Simulation results show that pseudo-second-order correlation coefficient (r₂² = 0.921) is higher than pseudo-first-order (r₁² = 0.7815) and intra-particle diffusion (r_i² = 0.905). Therefore, the adsorption process of CO₂ onto CaO impregnated AC filter media is appropriate for the pseudo-second-order kinetic model.

1. Introduction

Generally, people spend more than 90% of their time in indoor environments and therefore, there is concern regarding indoor pollutants [1]. Nowadays, buildings are being constructed with significantly stricter leak tightness requirements, as demanded by Building Regulations. There is increasing concern regarding emissions of CO₂ and the impact on health and well-being [2,3,4] and comfort [5,6] of occupants in air-tight housing. CO₂ is the representative pollutant of indoor air quality and its concentration is associated with human activity of about 650–700 ppm [7,8]. High concentrations of CO₂ are known to have various adverse effects such as headaches, drowsiness, and dizziness for residents [9,10,11]. Ventilation is one of the easiest ways to reduce CO₂ concentration, and the modern practice of completely insulating buildings to retain or send back heat also brings on reductions of IAQ (Indoor Air Quality).

Current and updated research on how to remove CO₂ effectively and economically from indoor air has become necessary. A few usual ways are being utilized to clean polluted air, including compromising adsorption, absorption, membrane, and cryogenic gas cleaning manners [12,13,14]. Adsorption has been shown to be a technique for conducting low concentrations of CO₂ [15,16,17] because of its simplification, low energy demands, and cost effectiveness [18]. Impregnated activated carbon as media for the sorption-type filter adsorption is another functional method for removal of CO₂ in a building [19]. Song et al. [16] investigated CaO modified silica adsorbents to have basic locations on their surfaces, and thus, have created a raised affinity for CO₂ molecules. Jensen et al. [20] carried out a comparable quantum chemical study of CO₂ adsorption on MgO and CaO, CO₂ adsorption onto CaO as single dentate geometry on both sides and angle sites. Lahuri et al. [21] evaluated CaO impregnated on iron (III) oxide bimetal adsorbent system as a possible source of base sites for CO₂ capture.

This study aims to examine the performance of CaO impregnated coconut shell AC adsorbent-loaded nonwoven fabric filter by adsorption characteristics, as well as the breakthrough curves and pressure drops under various testing conditions. Adsorption capacity and these characteristics were decided as the functional of CO₂ concentration and face velocities. We also examined adsorption kinetics, utilizing the pseudo-first-order, pseudo-second-order and intraparticle models and their kinetic constants, thus providing the basic data demanded for design and operation of equipment for air handling units of the building.

2. Experimental Method

Chemically modified activated carbon adsorbents were prepared by impregnation. Impregnation of CaO was performed as follows. Calcium acetate monohydrate (10 g) was mixed with 200 mL of deionized water by agitation for 5 h. Through vacuum and filtration processes remove undissolved salt to obtain saturated solution. One (1) g of support material was impregnated with this solution by agitation at 25 °C for 12 h, followed by suction filtration and drying at 80 °C for 12 h and then used vacuum drying at 120 °C for ~10 h. The impregnated support materials were finally calcined at 700 °C for 2 h by blowing rate of 1 L/min nitrogen.

Table 1. Procedures for preparation of impregnants.

Reagent and Amount	Solution (wt %)	DI Water (mL)	AC (g)
10 g Ca(CH3CO2)2·H2O	4.80%	200	10

summarizes the impregnation procedures employed.

Calcium acetate monohydrate was converted into calcium oxide through the following pathways during calcination.

Ca(CH₃CO₂)₂·H₂O → Ca(CH₃CO₂)₂ → CaCO₃ → CaO

Surface area, pore size distribution, and pore width of the completed sample sorbents were analyzed with an ASAP2020 (Micromeritics Instrument Corporation, GA, USA).

Figure 1 showed the schematic diagram of the experimental system. The CaO impregnated granular activated carbon (GAC)-loaded on nonwoven fabric filter media (supplied by AIRREX Co. Ltd., New Taipei City, Taiwan) was set in a designed 15 cm × 15 cm filter area. Two differential pressure gauges were utilized to monitor pressure drop before and after the filter. The testing rig was kept at 24 ± 1 °C temperature by air-conditioning environment control system in a cleanroom. The main testing airflow is from here which controlled at 22 ± 1 °C temperature and 40 ± 2% relative humidity. The face velocity of filter was measured and controlled from 0.3 m/s to 0.5 m/s (related with 0.7 to 1.0 m/s face velocity of a full-scale chemical filter actually operated) (the TSI 9535-A anemometer is ±3% of full-scale accuracy) with the invertor which connected to the three-phase air blower and flow damper. The compressed dry air with −40 °C dew point temperature passing through the impinger which produces the airflow with saturated contaminant (i.e., CO₂) then becomes the challenge gas entering the upstream air duct. The impinger, filled with 99.9% grade CO₂, is submerged in the brine water bath with adjustable water temperature from −15 °C to 25 °C. The upstream concentration in the testing rig is controlled by the flow rate passing thought the impinger, which is adjusted via a mass flow controller (LINTEC MC-700). The upstream concentrations were fixed at 800, 1000, and 1200 ppm with ±5% deviation. TES 1370 NDIR CO₂ Meter.

The testing method followed ASHRAE Standard 145.2 [23] and is similar to the researcher works [24,25]. As shown in Figure 2, conditioned air pass through the adsorbent. The upstream concentrations and downstream concentrations are simultaneously measured to decide removal efficiency (η) [26]. Breakthrough time (t_b) is defined as the time when the outlet concentration was 2% of the inlet concentration. Equilibrium time (t_t) is defined as the time when the outlet concentration was 98% of the inlet concentration.

The removal efficiency (η) of CO₂ decided and simultaneously monitored the upstream concentrations and downstream concentrations of test rig [26]:

$$Q=C_{0}V{\displaystyle \int \eta (t)dt\times \frac{1}{0.082}}\times \frac{1}{(273+k)}\times M\times {10}^{-6}$$

(1)

where Q is adsorption capacity; C₀ is the inlet concentration; V is the airflow rate; t is testing time; η is removal efficiency (%); and M is testing gas molecular weight.

3. Results and Discussion

3.1. Adsorption Capacity

Figure 3 presents CO₂ various adsorption capacity with various inlet concentrations and various face velocities. As shown in Figure 3, if the inlet concentration of the adsorbate is increased, resulting in increased diffusion velocity into the pores of the CaO impregnated AC filter, equilibrium adsorption may reach faster; the equilibrium time decreased from 237 to 86 min, 175 to 57 min and 168 to 49 min at 0.1, 0.2 and 0.3 m/s face velocities, respectively.

3.2. Breakthrough

Its determination was performed by various inlet concentration of CO₂ from 800 to 1200 ppmv at the face velocity of 0.1 to 0.3 m/s (Figure 4). As shown in Figure 4, increased the inlet concentration of the adsorbate, the breakthrough time was reduced. Furthermore, the face velocity was increased, and the breakthrough time also became shorter.

Nelson and Harder [27] developed the relationship between two different initial concentrations of denoted VOC (Volatile Organic Compound) in terms of the breakthrough time:

$$\frac{t_{b,1}}{t_{b,2}}={(\frac{C_{0,1}}{C_{0,2}})}^{\alpha }$$

(2)

where α is the average value of slopes of breakthrough time versus different initial CO₂ concentration curve plotted on logarithmic scales. We used it to predict the performance of CaO impregnated AC filter under indoor conditions. The relationship curves of different face velocities are shown in Figure 5 and

Table 2. Effect of initial concentrations on the breakthrough time.

Face Velocity (m/s)	Breakthrough Time vs. Initial Concentration
0.1	$t_b{C}_0^{0.4009}=15,060.037$, R2 = 0.9982
0.2	$t_b{C}_0^{0.3602}=29,584.327$, R2 = 0.9963
0.3	$t_b{C}_0^{0.3309}=60,596.96$, R2 = 0.9909

. From the data of Figure 5 and Table 2, we can conclude that once the slope of one breakthrough time–concentration relationship is known, the slope of the other breakthrough percentages can be approximated. If the breakthrough time at one concentration is known, breakthrough time at other concentrations can be calculated accordingly. Nevertheless, best results are obtained if each individual slope for a given set of conditions is determined experimentally.

3.3. Adsorption Kinetics

The kinetics of CO₂ adsorption onto CaO impregnated AC filter was investigated by using pseudo-first-order and pseudo-second-order models.

3.3.1. Pseudo-First-Order Model

The pseudo-first-order equation is given as [28]:

$$\log (q_e-q_t)=\log q_e-\frac{k_1}{2.303}t$$

(3)

Figure 6 presents log(q_e − q_t) against t of the pseudo-first order equation plots at the adsorption of CO₂. The sorption capacity, q_e,1, the first-order rate parameters, k₁, and correlation coefficients, r₁² are showed in

Table 3. Comparison of the pseudo-first- and second-order adsorption, and intraparticle diffusion rate constants.

Face Velocity	Inlet Concentration		Pseudo-First Order			Pseudo-Second Order			Intra-Particle Diffusion
		qe (exp, g)	k1 (min−1)	qe,1 (g)	r12	k2 (g−1 min−1)	qe,2 (g)	r22	ki (g min−1/2)	C	ri2
0.1 m/s	800 ppm	2.77	0.0082	2.73	0.7815	0.149	4.2	0.9519	0.3076	−0.4487	0.9935
	1000 ppm	3.66	0.0119	3.64	0.866	0.2783	2.51	0.9844	0.2153	−0.0979	0.9958
	1200 ppm	4.54	0.0217	4.51	0.7843	0.5714	1.24	0.9888	0.1397	−0.0266	0.9472
0.2 m/s	800 ppm	2.28	0.011	2.24	0.8213	0.0777	5.65	0.9504	0.5096	−1.0158	0.9811
	1000 ppm	3.01	0.0132	2.99	0.7827	0.1214	3.77	0.921	0.3727	−0.6231	0.9771
	1200 ppm	3.21	0.0324	3.16	0.8884	0.4141	1.58	0.9806	0.2182	−0.0286	0.9975
0.3 m/s	800 ppm	2.10	0.0108	2.10	0.7874	0.0496	6.96	0.9565	0.6473	−1.4821	0.9671
	1000 ppm	2.42	0.0214	2.41	0.8874	0.253	2.36	0.9857	0.7353	−0.5122	0.905
	1200 ppm	2.77	0.0337	2.75	0.9074	0.33	1.73	0.9847	0.2657	−0.1127	0.9888

. The q_e experimental values are also contained in Table 3 for comparison with those predicted. The equilibrium adsorption capacity of the experiment increased from 2.77 to 4.54, 2.28 to 3.21 and 2.1 to 2.77 mg g⁻¹ at face velocity 0.1, 0.2, and 0.3 m/s, respectively during CO₂ initial concentration increased from 800 to 1200 ppm, pointing out that CO₂ removal is relying on initial concentration. The adsorption capacity of the equilibrium increased from 2.73 to 4.51, 2.24 to 3.16 and 2.1 to 2.75 mg g⁻¹ at the initial concentration of CO₂ 800, 1000 and 1200 ppm, respectively when face velocity increased from 0.1 to 0.3 m/s, showing that the CO₂ removal is based on face velocity too. Also, q_e calculated values conform well with the experimental data. After all, k₁ rate constant values were discovered to increase from 0.0082 to 0.0217, 0.011 to 0.0324 and 0.0108 to 0.0337 min⁻¹ at face velocity 0.1, 0.2 and 0.3 m/s, respectively for an increase in the initial concentration from 800 to 1200 ppm. Since the adsorption kinetics follow pseudo-first-order, the rate constant k₁ values should increase linearly with increasing initial concentration [29,30].

3.3.2. Pseudo-Second-Order Model

The pseudo-second-order model is showed as: [31]

$$\frac{t}{q_t}=\frac{1}{k_2{q}_e^2}+\frac{1}{q_e}t$$

(4)

Figure 7 presents plots of t/q_t versus t of the pseudo-second order equation of CO₂ adsorption. The pseudo-second order rate parameters k₂ and the correlation coefficients r₂² are presented then compared with r₁², r_i², k₁, and k_i values for the pseudo-first order reaction mechanism and intraparticle model (Table 3). The adsorption capacity q_e,2 of the equilibrium increased from 0.309 to 0.498, 0.329 to 0.785, and 0.715 to 0.721 mg g⁻¹ at face velocity 0.1, 0.2, and 0.3 m/s, respectively. When CO₂ initial concentration increased from 800 to 1200 ppm, showed the initial concentration effect on CO₂ removal. The equilibrium adsorption capacity q_e,2 decreased from 4.2 to 1.24, 5.65 to 1.58, and 6.96 to 1.73 mg g⁻¹ at CO₂ initial concentration 800, 1000 and 1200 ppm, respectively when face velocity increased from 0.1 to 0.3 m/s, pointing out that the CO₂ removal is relying on face velocity too. k₂ rate constant values were discovered to increase from 0.149 to 0.5714, 0.0777 to 0.4141 and 0.0496 to 0.33 g mg⁻¹ min⁻¹ for an increase from the initial concentration 800 to 1200 ppm at face velocity 0.1, 0.2 and 0.3 m/s, respectively. Also, q_e calculated values conform well with the experimental data. The rate coefficient k₂ of the pseudo-second-order rate model is figured versus CO₂ initial concentration and is shown, the relation is not a simple function between k₂ and C₀ [29,30].

3.3.3. Intraparticle Diffusion Model

It was also shown that the intraparticle diffusion model [32,33] rate processes are generally presented in respect to square root of time. The following equation decided k_i rate parameters of intraparticle diffusion at various initial concentrations.

$$q_t=k_i{t}^{1/2}$$

(5)

As presented the intraparticle diffusion model at a face velocity of 0.1 m/s in Figure 8a, the external surface adsorption (Stage 1) is out. Before 16 min, Stage 1 is finished and then intraparticle diffusion control of Stage 2 is obtained and it goes on from 25 min to 196, 144 and 64 min at 800, 1000 and 1200 ppm inlet concentration, respectively. Lastly, Stage 3 equilibrium adsorption begins after 225, 169 and 100 min at the inlet concentration of 800, 1000 and 1200 ppm, respectively [34]. The CO₂ is slowly transferred with intraparticle diffusion into the particles and lastly stays in the micropores. Generally, intraparticle diffusion rate constant k_i is the slope of the line in Stage 2. Table 3 also listed the rate parameter k_i and its correlation coefficients. There was some control of boundary layer as it can be seen from the value of C.

4. Conclusions

The CaO impregnated AC filter system is suitable for reducing the CO₂ indoor air concentration in buildings with air condition system. In CO₂ single vapor adsorption measurements, the adsorption time decreased with inlet concentration of CO₂ and face velocity increased. The adsorption capacity increased with increased initial concentration and decreased face velocity, too. To increase the inlet concentration of the adsorbate, the breakthrough time was reduced. Furthermore the face velocity was increased, and the breakthrough time also became shorter. Once the slope of one breakthrough time–concentration relationship is known, the slope of the other breakthrough percentages can be approximated. The pseudo-first- and second-order kinetics, and intraparticle diffusion model also performed kinetic analysis for the adsorption of CO₂ onto CaO impregnated AC filter. The trend of adsorption of CO₂ onto CaO impregnated AC filter for various initial CO₂ concentrations over the complete range succeed the pseudo-second-order kinetic model of the test data fixed on the highest correlation coefficient of determination, R²(0.921) values which signifies a monolayer adsorption phenomenon exists between CaO impregnated AC filter and CO₂. The CO₂ is slowly transferred with intraparticle diffusion into the particles and is lastly kept in micropores.